[med-svn] [Git][med-team/bart][master] 636 commits: skip GPU tests when no GPU available
Martin Uecker (@uecker-guest)
gitlab at salsa.debian.org
Thu Dec 7 10:06:30 GMT 2023
Martin Uecker pushed to branch master at Debian Med / bart
Commits:
2fe678fa by Christian Holme at 2022-09-26T14:39:32+02:00
skip GPU tests when no GPU available
- - - - -
7dbdb62b by Christian Holme at 2022-09-26T15:16:10+02:00
.gitlab-ci: Unit Test LTO should run unit tests
and not the integration tests
- - - - -
b0c2c17f by Christian Holme at 2022-09-26T15:17:31+02:00
gitlab-ci: add LTO integration test
- - - - -
67c9dd27 by Christian Holme at 2022-09-27T10:11:03+02:00
gitlab-ci: put `if` and command on the same line
This way, the command being run can be seen in the
gitlab runner logs
- - - - -
cfa6a974 by Christian Holme at 2022-09-27T10:11:30+02:00
gitlab-ci: comment out LTO Integration Test
As it takes too long for now
- - - - -
809d310b by Martin Uecker at 2022-10-02T17:34:03+02:00
remove unsigned (multind)
- - - - -
c16bfc83 by Martin Uecker at 2022-10-02T17:34:03+02:00
relax two tests
- - - - -
9114f1d7 by Martin Uecker at 2022-10-02T17:38:19+02:00
_Bool -> bool
- - - - -
cf5036a3 by Martin Uecker at 2022-10-02T17:40:19+02:00
flags are unsigned long
- - - - -
0b430892 by Martin Uecker at 2022-10-02T18:33:27+02:00
unsigned flpmath
- - - - -
2eba55ce by Martin Uecker at 2022-10-02T18:41:54+02:00
flags are unsigned long (flpmath.h)
- - - - -
e521f64a by Martin Uecker at 2022-10-02T18:43:49+02:00
_Bool -> bool
- - - - -
a686f3b0 by Martin Uecker at 2022-10-02T21:48:54+02:00
missing new line
- - - - -
5428c0ae by Martin Uecker at 2022-10-02T21:48:54+02:00
relax test
- - - - -
3a84b390 by Martin Uecker at 2022-10-06T15:25:13+02:00
minor
- - - - -
1d227f55 by Martin Uecker at 2022-10-06T15:25:16+02:00
reporting problems
- - - - -
9dcaac59 by Martin Uecker at 2022-10-06T15:25:16+02:00
remove old cmake file
- - - - -
efb8777e by Martin Uecker at 2022-10-11T14:28:12+02:00
fix exponent
- - - - -
1ead7298 by Martin Uecker at 2022-10-11T14:33:32+02:00
rewrite Si
- - - - -
1720eeee by Christian Holme at 2022-10-11T14:51:58+02:00
tests/rtnlinv: add dimensions tests for rt-ENLIVE
- - - - -
af4464c3 by Christian Holme at 2022-10-11T14:51:58+02:00
rtnlinv.c: fix ENLIVE dims
Somewhere, the dimension logic was messed up...
- - - - -
4529e45d by Martin Uecker at 2022-10-11T15:40:12+02:00
updates
- - - - -
7f5af323 by Martin Uecker at 2022-10-11T15:51:10+02:00
horner schema
- - - - -
2d4ba14c by Martin Uecker at 2022-10-13T19:08:59+02:00
pow -> powf
- - - - -
e5f47794 by Martin Uecker at 2022-10-13T19:10:18+02:00
avoid overflowing factorial
- - - - -
e8aa9da6 by Martin Uecker at 2022-10-13T19:23:22+02:00
copyright
- - - - -
62bddb12 by Martin Uecker at 2022-10-14T13:19:19+02:00
fix E-notation
- - - - -
97608a7e by Martin Uecker at 2022-10-29T15:28:52+02:00
unsigned flags
- - - - -
bb37b37c by Martin Uecker at 2022-10-29T15:28:52+02:00
make some small dims signed
- - - - -
694f25ad by Martin Uecker at 2022-10-29T16:02:05+02:00
remove unsigned from dim
- - - - -
59c765a6 by Moritz Blumenthal at 2022-10-31T13:13:20+01:00
fix linop_stack
- - - - -
949c32ce by Moritz Blumenthal at 2022-10-31T13:13:20+01:00
fix memleak
- - - - -
75a64c2c by Moritz Blumenthal at 2022-10-31T13:13:20+01:00
fix memleak
- - - - -
3188109b by Moritz Blumenthal at 2022-10-31T13:13:20+01:00
fix nlcmp
- - - - -
f9a4e962 by Moritz Blumenthal at 2022-10-31T13:13:20+01:00
fix real in t1fun
- - - - -
ec4d14fb by Moritz Blumenthal at 2022-10-31T13:13:20+01:00
fix UB
- - - - -
f151f2ae by Moritz Blumenthal at 2022-10-31T13:14:48+01:00
fixup unsigned for flpmath in utest
- - - - -
f99589ef by Moritz Blumenthal at 2022-10-31T13:15:20+01:00
unsigned int in fft-cuda
- - - - -
95d46730 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
fix psf for basis with higher dims
- - - - -
df91b490 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
update lists
- - - - -
add698ec by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
Add linop is identity
- - - - -
f411f5aa by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
get list in attach
- - - - -
e9850311 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
stream selection for gpukernel
- - - - -
46039c65 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
Prepare gpu init for entering omp regions
- - - - -
64a18ee1 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
rename zrblock->rblock and zrdiag->rdiag
- - - - -
04ad6c43 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
add nlops with jacobian for wirtinger calculus
- - - - -
b0e71e65 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
speedup nlop_zrprecomp
- - - - -
083c0269 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
Add linop_flip
- - - - -
46326c01 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
update someops
- - - - -
0367b820 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
refactor and add activations
- - - - -
4529ef05 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
add cardioid activation as option to resnet
- - - - -
fe70cda0 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
add function to compress binary floats
- - - - -
29885165 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
reduce memory for relu
- - - - -
7ccd3825 by Moritz Blumenthal at 2022-10-31T13:18:54+01:00
graph optimization remove constant nodes
- - - - -
21bd1db2 by Moritz Blumenthal at 2022-10-31T13:39:56+01:00
twixread fix for linectr/partctr
- - - - -
f9f3fc9c by Moritz Blumenthal at 2022-10-31T13:39:56+01:00
option to turn of consistency check
- - - - -
8fa58b26 by Moritz Blumenthal at 2022-10-31T13:39:56+01:00
add dimensions
- - - - -
f2aaa8b0 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
meaningful error if not compiled with TensorFlow
- - - - -
7f7f535a by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
multiple outputs for tf wrapper and detection of inputs and outputs
- - - - -
d7bc24be by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
reuse tf graph
- - - - -
023e1f69 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
set tf options (threads/memory/selected gpu)
- - - - -
b37fead6 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
remove unused code
- - - - -
bf67f0ea by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
get args from shared graph
- - - - -
1870e77e by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
allways pass grad_ys (tf2)
- - - - -
50c99b99 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
support for tensorflow 2 saved model
- - - - -
ce48c61a by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
fix ifdefs for tensorflow
- - - - -
4edbe638 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
checkpointing for tensorflow wrapper
- - - - -
4b299a75 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
fix warnings
- - - - -
69301151 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
reduce logging for tensorflow
- - - - -
81a0ed89 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
update tf_wrapper
- - - - -
4558bfd7 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
tf wrapper restructure and allocate tensors of different types
- - - - -
effd0200 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
Support for complex tensors in tensorflow wrapper
- - - - -
4ccf0494 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
add support for tensorflow graphs with jacobian
- - - - -
dff91b32 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
add utests for tensorflow wrapper
- - - - -
01058c71 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
update documentation
- - - - -
8b30558e by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
Initialize with constant array
- - - - -
2885cc25 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
move residual to interface
- - - - -
393233ff by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
move bart to channel first dims to networks
- - - - -
c05a8d63 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
add tensorflow graph to reconet
- - - - -
ad43b069 by Moritz Blumenthal at 2022-10-31T13:57:36+01:00
fix dependencies
- - - - -
e4c25fdc by Martin Uecker at 2022-10-31T14:48:52+01:00
coding style
- - - - -
03c98a03 by Moritz Blumenthal at 2022-10-31T15:33:55+01:00
optimize flattening
- - - - -
5baa67de by Moritz Blumenthal at 2022-10-31T15:33:55+01:00
join frw and adj
- - - - -
a553ef05 by Moritz Blumenthal at 2022-10-31T15:33:55+01:00
move rvc in model
- - - - -
ac174b82 by Moritz Blumenthal at 2022-10-31T15:33:55+01:00
automatic optimization in noir model
- - - - -
f7faa384 by Moritz Blumenthal at 2022-10-31T15:33:55+01:00
apply mask on image, not coil image
- - - - -
d371a771 by Moritz Blumenthal at 2022-11-01T15:12:44+01:00
cache nufft for gridding
- - - - -
3b29e6d7 by Moritz Blumenthal at 2022-11-01T15:20:37+01:00
make lowmem gridding independent
- - - - -
d75761ff by Moritz Blumenthal at 2022-11-01T15:24:11+01:00
update lowmem code
- - - - -
0b272fd2 by Moritz Blumenthal at 2022-11-02T14:38:46+01:00
refactor linop_loop (include copy wrapper)
- - - - -
4e32075b by Moritz Blumenthal at 2022-11-02T14:39:02+01:00
reduce temporary memory
- - - - -
353e20f7 by Moritz Blumenthal at 2022-11-02T14:39:23+01:00
parallelize gridding over outer dims
- - - - -
263fd6b4 by Moritz Blumenthal at 2022-11-03T17:38:52+01:00
add function to directly manage pointer by multiplace
- - - - -
f2a1d183 by Moritz Blumenthal at 2022-11-03T17:40:33+01:00
reduce peak memory in nufft
- - - - -
34c85e3f by Moritz Blumenthal at 2022-11-03T17:45:52+01:00
turn off caching of psf_nufft by default
- - - - -
2695b63d by Nick Scholand at 2022-11-04T15:37:06+01:00
Fix coeff split in poly kernel
- - - - -
5fbc01e1 by Nick Scholand at 2022-11-04T15:40:49+01:00
Simplify sampling
- - - - -
7d459b6a by Nick Scholand at 2022-11-04T15:40:59+01:00
Make BART geom accessible for component split with test
- - - - -
c938d542 by Nick Scholand at 2022-11-04T15:43:21+01:00
Combine individual components of BART logo
- - - - -
e7c3909a by Martin Uecker at 2022-11-06T14:11:35+01:00
fix sign of flags
- - - - -
67941a77 by Martin Uecker at 2022-11-06T14:11:35+01:00
more sign changes
- - - - -
7721f65d by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
fix warning
- - - - -
a69889e0 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
toeplitz_lowmem: inplace mull with psf if possible
- - - - -
c4265418 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
split lowmem code in extra functions
- - - - -
8bb29d36 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
make grid.h c++ compatible
- - - - -
22885212 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
gpu misc for optimal block/grid size
- - - - -
d792d998 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
make device functions static
- - - - -
83f505b3 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
optimize fftmod for 2d/3d continous strides
- - - - -
2f2c01c7 by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
add function to apply linphase
- - - - -
812a2a8b by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
Add function to apply rolloff correction
- - - - -
1326cd8d by Moritz Blumenthal at 2022-11-06T14:15:56+01:00
nufft add lowmem option which does not precompute anything
- - - - -
988ff35f by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
Chaining of nufft normal by linop framework
... this reduces the overhead as not src, dst and tmp are not
allocated at the same time.
- - - - -
12b5a260 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
Don't copy sens map in linop but use existing map
- - - - -
d660323c by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
Add operator stack linop output and dup input
- - - - -
6266e910 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
Option to stack (non-cart) SENSE model
... for a very low memory overhead
- - - - -
b569300a by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
use fftuc instead of linop (no tmp memory)
- - - - -
adb99d49 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
lowmem stacking along arbitrary dimensions
- - - - -
707a4e08 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
add lowmem tests for pics
- - - - -
32a3e3c6 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
more efficient shuffling on gpu
- - - - -
651fd7e3 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
gpu gridding
- - - - -
54a2c6f9 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
add gpu support to adjoint/forward nufft
- - - - -
aa8f7ab2 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
make nvcc overwritable
- - - - -
bae0825b by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
gpu use nested lock
- - - - -
fe1bf974 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
complete reco on GPU
- - - - -
c3f81320 by Moritz Blumenthal at 2022-11-06T14:15:57+01:00
relax test for gwdg pipeline
- - - - -
471ae4d2 by Martin Uecker at 2022-11-06T15:02:11+01:00
fix memory leak
- - - - -
2785e416 by Martin Uecker at 2022-11-06T15:02:23+01:00
fix typo
- - - - -
6dcd691e by Christian Holme at 2022-11-07T12:06:55+01:00
gridding: error on incompatible dims
- - - - -
f8124338 by Moritz Blumenthal at 2022-11-07T16:21:53+01:00
turn off gridding on GPU by default
- - - - -
ed112e5b by Philipp Ehses at 2022-11-09T19:00:55+01:00
bart.py: Print stdout and stderr output from bart call
Previously, it was difficult to get access to bart's error codes and
messages from python, which is especially relevant for interactive
sessions. This patch addresses this issue.
Stdout, stderr, and the error code are stored as function attributes,
so that they can be accessed from outside of bart.py.
Example:
from bart import bart
bart(0, "pics -h")
print("Here's the command output again:")
print(bart.stdout)
- - - - -
e394ee02 by Philipp Ehses at 2022-11-09T19:00:55+01:00
thread-safety: do not use function attributes within the function
store them for debugging purposes only
- - - - -
c777c22d by Moritz Blumenthal at 2022-11-11T10:17:03+01:00
fix fftmod on gpu and add tests
- - - - -
3fcb897f by Christian Holme at 2022-11-11T10:20:20+01:00
fixup: re-add CUDA fft unit test
- - - - -
11a8a78d by Christian Holme at 2022-11-11T11:21:24+01:00
tests/fft: fix fftmod vs fftshift test
- - - - -
4ded5814 by Alexander Fyrdahl at 2022-11-15T21:38:25+01:00
fix kwargs
- - - - -
8c851ae4 by Moritz Blumenthal at 2022-11-24T14:18:56+01:00
Add cpu wrapper for operators
- - - - -
dfb4fa9e by Moritz Blumenthal at 2022-11-25T11:58:43+01:00
Add gpu support for lowrank (with CPU prox)
- - - - -
76c4acae by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
Relax lowmem tests
CG piles up "large" errors from small floating point deviations,
hence, we test after 2 iterations with a sharp limit and after
30 iterations with a loose limit
- - - - -
aae84798 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
fix c++ compatibility of multiplace.h
- - - - -
2ab21d21 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
multiplace_move_F for gpu
- - - - -
ba378629 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
refactor initialization of kb_table
- - - - -
b00a4022 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
fix scaling of nufft
- - - - -
538f6a97 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
tighten test for fixed scaling
- - - - -
005c1985 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
optimize linop_fftc
- - - - -
b09bd4a7 by Moritz Blumenthal at 2022-11-29T10:05:10+01:00
configure fftw wisdom/measure via environment variable
- - - - -
23ca95c1 by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
optimize fftmod for dims which are multiple of 4
- - - - -
e8c89d67 by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
make defintion of rolloff consistent
- - - - -
ef289f40 by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
extend domain of ftkb
- - - - -
d7a26a6f by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
grid decomposed in normal mode
- - - - -
f9e6f770 by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
more efficient parallel loop
- - - - -
626a0cd0 by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
more efficient linphases
- - - - -
4e7f3f99 by Moritz Blumenthal at 2022-12-01T10:41:35+01:00
compute psf on decomposed grid
- - - - -
6b8785a8 by Christian Holme at 2022-12-01T10:42:50+01:00
num/ops.c: make gpu-wrapper spam less
- - - - -
fcb7a5de by Christian Holme at 2022-12-01T10:42:50+01:00
nrmse: add option for scientific notation
- - - - -
af3cd75b by Christian Holme at 2022-12-01T10:42:50+01:00
rtnlinv: coding style and less debug spam
- - - - -
879ae2db by Christian Holme at 2022-12-01T10:42:50+01:00
nlinv: add option to specifiy non-Cart image size
but unlike rtnlinv, we specify output image size
and not the oversampled size
- - - - -
e9874fd4 by Christian Holme at 2022-12-01T10:42:50+01:00
nufft: use -x for image dimensions
as -d is often used for debug level. But we keep -d around for
compatibility, for now
- - - - -
49354289 by Christian Holme at 2022-12-01T10:42:50+01:00
nufft: document maxiter for inverse nuFFT
- - - - -
f786a48f by Martin Uecker at 2022-12-03T21:25:54+01:00
fix warning
- - - - -
f43b670b by Martin Uecker at 2022-12-08T17:41:28+01:00
coding style
- - - - -
41d2299e by Martin Uecker at 2022-12-08T17:56:25+01:00
coding style
- - - - -
3a3bea8d by Martin Uecker at 2022-12-08T18:08:23+01:00
reduce noise, and tabs
- - - - -
671e6501 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
fix normal of linop_stack_cod for inplace
- - - - -
b2039d96 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
add maxeigen scaling at samplace
- - - - -
fb20cd56 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
Multiple reularizations for primal dual
- - - - -
889c4e29 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
maxeigen scaling in fista/pridu
- - - - -
fc092c63 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
long dims support
- - - - -
731b41aa by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
multiplace for preconditioning
- - - - -
d4f22c17 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
Add script to compute k-space preconditioner
- - - - -
81b07593 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
prox for weighted least squares + gpu support
- - - - -
7d651780 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
add function for preconditioning
- - - - -
632fa98e by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
preconditioning in pics
- - - - -
b0fb68de by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
reduce tmp memory in ops
- - - - -
db485f53 by Moritz Blumenthal at 2022-12-14T10:06:53+01:00
nufft inverse gridding on gpu
- - - - -
9e75c038 by Moritz Blumenthal at 2022-12-15T09:39:04+01:00
reduce peak memory in chambolle pock
- - - - -
226258a1 by Moritz Blumenthal at 2022-12-15T09:39:04+01:00
add subspace to nufft
- - - - -
dcc9c2af by Moritz Blumenthal at 2022-12-15T09:39:04+01:00
add subspace to kspace preconditioner
- - - - -
bd18a37a by Christian Holme at 2023-01-05T15:09:25+01:00
tests: more robust traj rot test
For example on riscv64, it would output a shift of 29.999998
instead of 30.00000, failing the test.
Now we check the nrmse by writing it to a file.
- - - - -
5338ee1a by Moritz Blumenthal at 2023-01-05T15:36:31+01:00
fix old scaling for compatibility mode
- - - - -
cd974af5 by Moritz Blumenthal at 2023-01-05T15:36:48+01:00
fix nufft psf for dims not a multiple of 4
- - - - -
3deb91bf by Christian Holme at 2023-01-05T16:06:12+01:00
pics: add option to specify FISTA
- - - - -
a6e0c6ad by Christian Holme at 2023-01-05T16:18:59+01:00
noncart gpu pics test: use FISTA instead of CG
As this works with a much smaller tolerance
- - - - -
ba80d973 by Moritz Blumenthal at 2023-01-19T11:11:16+01:00
int -> long in gpukrnls
- - - - -
b8e75463 by Martin Uecker at 2023-01-30T23:27:17+01:00
missing spaces
- - - - -
60df0baf by Martin Uecker at 2023-01-30T23:27:17+01:00
VLA macro is only for headers
- - - - -
653f2a7b by Martin Uecker at 2023-01-30T23:31:10+01:00
indent switch correctly
- - - - -
4c01b8f5 by Martin Uecker at 2023-01-31T22:18:57+01:00
spaces...
- - - - -
26190165 by Martin Uecker at 2023-02-01T22:53:03+01:00
unsigned removal
- - - - -
c362c0a9 by Martin Uecker at 2023-02-05T10:11:05+01:00
space
- - - - -
5f7d8a76 by Martin Uecker at 2023-02-05T10:40:12+01:00
remove noir_create{2,3}
- - - - -
42f81936 by Martin Uecker at 2023-02-05T13:24:36+01:00
remove unused headers
- - - - -
aac8a228 by Martin Uecker at 2023-02-07T21:50:53+01:00
remove unsigned from ode
- - - - -
1537e9b8 by Martin Uecker at 2023-02-09T00:35:53+01:00
move perror
- - - - -
039fc790 by Martin Uecker at 2023-02-10T08:49:10+01:00
change default permission for new files
- - - - -
79ddde78 by Martin Uecker at 2023-02-12T20:57:32+01:00
remove unsigned
- - - - -
5947a32f by Martin Uecker at 2023-02-12T20:58:09+01:00
fix mismatching bounds
- - - - -
ae6091de by Martin Uecker at 2023-02-28T20:44:18+01:00
fix linop_transpose_adjoint for non-squared matrices
- - - - -
e84f4ba7 by Christian Holme at 2023-03-01T16:16:20+01:00
pics: merge options to select algorithms
- - - - -
af1c1b42 by Martin Uecker at 2023-03-11T17:52:31+01:00
minor
- - - - -
cc567740 by Martin Uecker at 2023-03-11T17:52:31+01:00
adjoint state
- - - - -
46f30ed1 by Martin Uecker at 2023-03-11T17:52:31+01:00
use nested functions
- - - - -
d459ffe7 by Martin Uecker at 2023-03-11T17:52:31+01:00
quadrature
- - - - -
f862d2fd by Martin Uecker at 2023-03-11T17:52:31+01:00
clang fixes
- - - - -
646de6b0 by Christian Holme at 2023-03-14T13:46:41+01:00
fix minor issues
Resolves #300, resolves #301
- - - - -
93b7c451 by Martin Uecker at 2023-03-14T22:41:41+01:00
coding style / comparison of angles
- - - - -
13b57d54 by Martin Uecker at 2023-03-14T22:41:41+01:00
move sinc into specfun
- - - - -
0f27299a by Martin Uecker at 2023-03-14T22:41:41+01:00
minor
- - - - -
c7bd1708 by Martin Uecker at 2023-03-14T22:41:41+01:00
move matrix ops to numerical libary
- - - - -
17c87496 by Martin Uecker at 2023-03-14T22:41:41+01:00
use quadrature
- - - - -
9cbcea5d by Martin Uecker at 2023-03-14T22:41:41+01:00
simplify
- - - - -
ade93cc5 by Martin Uecker at 2023-03-14T22:41:41+01:00
simplify sinc pulse mathematics
- - - - -
df5b98f9 by Martin Uecker at 2023-03-14T22:41:41+01:00
improve unit test
- - - - -
ff14fb8b by Martin Uecker at 2023-03-14T22:41:41+01:00
turn off contracting of floating point ops
- - - - -
9c6c3213 by Martin Uecker at 2023-03-14T22:41:41+01:00
relax some tests for Macs with M1
- - - - -
e776766b by Martin Uecker at 2023-03-15T08:10:39+01:00
relax one more
- - - - -
f3baa2b4 by Nick Scholand at 2023-03-15T15:49:19+01:00
Option to pass ODE and STM tolerance to sim and moba
- - - - -
68990787 by Nick Scholand at 2023-03-15T15:49:19+01:00
Dynamic rephaser
- - - - -
b96e1bd3 by Nick Scholand at 2023-03-15T15:49:19+01:00
Fix sobolev norm for Bloch moba
- l2 penalty on B1
- regularization towards 0
- - - - -
751cd243 by Nick Scholand at 2023-03-15T15:49:19+01:00
Fix sobolev norm for reparameterized LL model
- Add l2 norm to map
- Fix FFT in initialization
- - - - -
6fae99bb by Nick Scholand at 2023-03-15T15:49:19+01:00
Option to change sobolev norm strength on B1/R1' map
- - - - -
d7a66806 by Nick Scholand at 2023-03-15T15:49:19+01:00
fix r2 scaling
- - - - -
5e63262f by Nick Scholand at 2023-03-15T15:49:19+01:00
comments
- - - - -
a4e4c8a6 by Nick Scholand at 2023-03-15T15:49:19+01:00
Pass B0 map to moba with tests
- - - - -
105a2de5 by Nick Scholand at 2023-03-15T15:49:19+01:00
Make regularization accessible from command line
- - - - -
ade8da90 by Nick Scholand at 2023-03-15T15:49:19+01:00
minor
- - - - -
333f7789 by Nick Scholand at 2023-03-15T15:49:19+01:00
Add partial update to irgnm2
with (1-ratio) contribution of the solution of the previous step
- - - - -
209a4d8b by Nick Scholand at 2023-03-15T15:49:19+01:00
Make l2 flag on parameter maps accessible from CLI
- - - - -
c791e76b by Nick Scholand at 2023-03-15T15:49:19+01:00
Make wavelet strength accessible for moba CL tool
- - - - -
26bff339 by Nick Scholand at 2023-03-15T15:49:19+01:00
Accessible scaling for T1phy + Regularization of R1' around 0
- - - - -
5d367184 by Nick Scholand at 2023-03-15T15:49:19+01:00
Extend accessibility to all models
- - - - -
e3978a1f by Nick Scholand at 2023-03-15T15:49:19+01:00
comment
\
- - - - -
2a5d33ef by Nick Scholand at 2023-03-15T15:49:19+01:00
Input ksp sensitivities to moba
- - - - -
19f5d3e4 by Nick Scholand at 2023-03-15T15:49:19+01:00
Export ksp coils in moba
- - - - -
e4a8aab0 by Nick Scholand at 2023-03-15T15:49:19+01:00
Option to turn off l2 on coils in moba
- - - - -
c69ac177 by Nick Scholand at 2023-03-15T15:49:19+01:00
Option to turn off coil updates in moba
- - - - -
18287d1c by Martin Uecker at 2023-03-15T15:49:19+01:00
coding style
- - - - -
ce15aec0 by Martin Uecker at 2023-03-15T15:49:19+01:00
remove unneeded variable
- - - - -
8fdf64ce by Martin Uecker at 2023-03-15T18:23:26+01:00
move partial update into callback
- - - - -
7a90f1e9 by Martin Uecker at 2023-03-15T18:23:26+01:00
coding style
- - - - -
06d1f2a8 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
make ref/unref of shared pointers threadsafe
- - - - -
88f1d03d by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
Turn off TensorCores in cuDNN 8
- - - - -
f2a2af29 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
Optimizations for cuDNN
- - - - -
a714b056 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
multigpu and streams for cudafft
- - - - -
d962aece by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
threadsafe cuda plan creation
- - - - -
744c1539 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
Add AVL tree (threadsafe)
- - - - -
3e99a311 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
Implement memcache by binary search tree
- - - - -
ed6d3788 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
Add utest for memcache
- - - - -
43c48a7d by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
add functions to manage/sync a pool of cuda streams
- - - - -
aac89ef8 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
reinitialize gpu when entering omp region
- - - - -
d4786869 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
use initialized number of streams in op_loop
- - - - -
6a10b87b by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
updates for managed memory and cuda_get_device num
- - - - -
7b821236 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
merge gpu and copy wraper
- - - - -
a5f32ddd by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
multigpu for multiplace
- - - - -
f2cd0854 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
multigpu for maps operator
- - - - -
389af35b by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
make multiplace threadsafe
- - - - -
f55b3a86 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
pics multigpu now works
- - - - -
e8740fdf by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
add function to assign op to specific gpu
- - - - -
f1656e59 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
add function to assign linop to specific gpu
- - - - -
94294852 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
add function to assign nlop to specific gpu
- - - - -
872009dd by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
add function to assign nn to specific gpu
- - - - -
92cc75fe by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
update checkpointing
- - - - -
80c5bdaa by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
add multigpu to stacking container
- - - - -
d516d47a by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
multigpu with nlop stack container
- - - - -
6a72f007 by Moritz Blumenthal at 2023-03-15T18:23:26+01:00
nn multigpu stack
- - - - -
1d273c30 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
multigpu for reconet
- - - - -
610f5819 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
performance for scalar ops on gpu
- - - - -
251bfb03 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
update blas md_wrapper
- - - - -
e6181626 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
updates for vecops_strided
- - - - -
f39cae01 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
refactor gpumisc
- - - - -
bc144204 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
use gpukrnls_misc in gpu_conv
- - - - -
ab680f4f by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
Add kernels for unfolding
- - - - -
a2d10fd3 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
simplify batchnorm/bias (use optimized strided calls)
- - - - -
017fe813 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
make all blas calls threadsafe by omp lock
- - - - -
a5fe94ac by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
debugging/printing for strided vecops
- - - - -
d179a4ce by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
set threads for nested parallelism
- - - - -
634f7ff6 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
fix GPU_ASSERTS
- - - - -
e308b0b0 by Moritz Blumenthal at 2023-03-15T18:23:27+01:00
kernel for strided copies
- - - - -
83529cc5 by Martin Uecker at 2023-03-15T20:49:54+01:00
relax one test
- - - - -
0622c290 by Martin Uecker at 2023-03-16T16:53:33+01:00
white space, copyright, style
- - - - -
2af3986c by Martin Uecker at 2023-03-16T16:53:33+01:00
simplify rules for header, git tracks it for us
- - - - -
7f84ac10 by Martin Uecker at 2023-03-17T16:06:30+01:00
do not use -ffp-contract=off
- - - - -
0313f4d0 by Martin Uecker at 2023-03-17T16:56:43+01:00
new complex sq kernel
- - - - -
da71aa4d by Nick Scholand at 2023-03-19T15:28:07+01:00
Remove warning and fix nan issue in magnetization preparation
- - - - -
017b8e56 by Nick Scholand at 2023-03-19T18:34:35+01:00
Single regularization for T1phy model in moba
Modified regularization for FA=1 requires test to be initialized closer
to solution.
- - - - -
e3a8b577 by Nick Scholand at 2023-03-19T18:34:35+01:00
Separate function for readout relaxation rate conversion
- - - - -
c5b4fcd2 by Nick Scholand at 2023-03-19T18:34:35+01:00
Replace by macro
- - - - -
a8123d34 by Martin Uecker at 2023-03-19T18:38:11+01:00
move macro
- - - - -
02f368ee by Christian Holme at 2023-03-20T16:28:55+01:00
tests: only compile needed commands
- - - - -
e1e0b525 by Christian Holme at 2023-03-20T16:41:33+01:00
tests/noise: add test for spike noise
- - - - -
0c847a24 by Nick Scholand at 2023-03-27T20:17:19+02:00
Fix special case with ROT test
Special case: TE==TR or TE==TRF together with slice-selection gradient
for ROT simulation
- - - - -
ef982b56 by Moritz Blumenthal at 2023-03-29T20:40:13+02:00
allow inplace for identity
- - - - -
ee7ffb31 by Moritz Blumenthal at 2023-03-29T20:40:53+02:00
use normal operator in irgnm
- - - - -
f6df6e78 by Moritz Blumenthal at 2023-03-29T20:46:20+02:00
don't perform clear in resize if not necessary
- - - - -
a652319b by Moritz Blumenthal at 2023-03-29T20:46:52+02:00
add functions for ops to directly free
- - - - -
e605fd65 by Moritz Blumenthal at 2023-03-29T20:47:23+02:00
optimize operator graph with linop identities
- - - - -
5e22e2d9 by Moritz Blumenthal at 2023-03-29T20:47:58+02:00
optimize nlop_flatten
- - - - -
fa776df4 by Moritz Blumenthal at 2023-03-29T20:48:47+02:00
extend/optimize linops
- - - - -
34749a7c by Moritz Blumenthal at 2023-03-29T20:49:39+02:00
Add function to merge cdiag and fftmod
- - - - -
b99e37d6 by Moritz Blumenthal at 2023-03-29T20:50:18+02:00
add function to construct sobolev operator
- - - - -
95153a58 by Martin Uecker at 2023-03-29T21:00:23+02:00
white space, style
- - - - -
3607e734 by Martin Uecker at 2023-03-29T21:26:35+02:00
options to suppress certain GCC 11 warnings for now
- - - - -
29cc683e by Martin Uecker at 2023-03-31T14:59:27+02:00
update README
- - - - -
5a5c11ed by Moritz Blumenthal at 2023-03-31T21:18:11+02:00
add test for sms nlinv noncart
- - - - -
c630f485 by Martin Uecker at 2023-04-08T13:16:09+02:00
nonnull fixes
- - - - -
f405f63d by Martin Uecker at 2023-04-16T20:10:40+02:00
coding style (Closes: #303, Closes #304)
- - - - -
3664133f by Alexander Fyrdahl at 2023-04-21T22:59:43+02:00
Incorrect grouping of expressions
- - - - -
2b414cdc by Martin Uecker at 2023-04-27T19:06:22+02:00
remove unsigned
- - - - -
fe9a74be by Martin Uecker at 2023-04-30T13:49:07+02:00
use xmalloc
- - - - -
1fb8e924 by Martin Uecker at 2023-04-30T19:09:42+02:00
avoid potential null derefence
- - - - -
4db1ec0d by Martin Uecker at 2023-04-30T19:10:13+02:00
address some analyzer warnings
- - - - -
499c4f21 by Martin Uecker at 2023-04-30T19:10:26+02:00
do not call snprintf with null
- - - - -
4e046ea7 by Martin Uecker at 2023-04-30T20:15:45+02:00
cleanup + analyzer workaround
- - - - -
a88ec880 by Martin Uecker at 2023-04-30T20:25:48+02:00
analyzer built
- - - - -
07d0e919 by Martin Uecker at 2023-05-19T17:39:12+02:00
coding style
- - - - -
894b1f65 by Martin Uecker at 2023-05-19T17:39:12+02:00
remove some more unsigned
- - - - -
31dd0409 by Martin Uecker at 2023-05-19T17:39:12+02:00
another quicksort test
- - - - -
4f508c1c by Nick Scholand at 2023-05-31T18:09:38+02:00
Constrain individual components to 1
Simulation requires components with default amplitude 1
- - - - -
77b04065 by Nick Scholand at 2023-05-31T18:09:38+02:00
Add BRAIN geometry to phantom tool with tests
- - - - -
41029912 by Martin Heide at 2023-05-31T18:09:38+02:00
added 64ch sensitivities
- - - - -
9284be07 by martin.schilling at 2023-05-31T18:09:38+02:00
Add tool for extracting spline control points from SVG files
- - - - -
d7d2234f by Nick Scholand at 2023-05-31T18:09:38+02:00
Add multi cfl passing option for geometry to phantom tool with tests
- - - - -
69b72d86 by Nick Scholand at 2023-05-31T18:09:38+02:00
Change 3D sens interface
- Allows for using less than 8 3D sensitivities
- Introduces struct for future extensions of phantom tool
- Add 3D sens support for polygon based phantoms
- - - - -
2d44a319 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
zero-overhead mode for nuFFT
- - - - -
6bd8f30d by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
optimize toeplitzmult for non-basis case
- - - - -
6ad67ad9 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
Fix race condition in cufft
- - - - -
053032cd by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
rolloff with strides
- - - - -
46ba94c3 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
double precision for precomputed bessel_beta
- - - - -
3084c814 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
double precision bessel function
- - - - -
2a216ce0 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
make width/os config of nufft
- - - - -
0581e208 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
simple chained linop for nufft supporting oversampling other than 2
- - - - -
2bcae977 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
set trajectory of reshaped nufft
- - - - -
ee39238b by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
add functions to move psf kern into nufft
- - - - -
4a7983d4 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
move psf_kern into nufft
- - - - -
e7679c65 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
move psf kern into nufft for decomposed gridding
- - - - -
448cce41 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
remove unused kern functions
- - - - -
3295c484 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
grid with strides
- - - - -
7793d4a7 by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
also loop over singleton ksp_dims
- - - - -
d198fe0c by Moritz Blumenthal at 2023-06-01T16:19:31+02:00
merge dims in grid for efficient subspace gridding
- - - - -
9fc8fcd3 by Moritz BLumenthal at 2023-06-01T16:19:31+02:00
merge apply linphase with fftmod and increase precision
- - - - -
95cf1bbe by Moritz Blumenthal at 2023-06-07T18:36:57+02:00
tool to compute basis function for nufft
- - - - -
19bc5ffb by Christian Holme at 2023-06-13T15:41:05+02:00
gitlab-ci: updates for new debian release
* new versions of clang and libgfortran
* use explicit debian codenames instead of "stable" and "testing"
- - - - -
a9aee034 by Martin Uecker at 2023-06-14T16:45:41+02:00
add error for adjoint + inverse in nufft
- - - - -
099af404 by Martin Uecker at 2023-06-14T16:45:41+02:00
rotation for Cartesian trajectories
- - - - -
fd8b2a4d by Martin Uecker at 2023-06-14T16:45:41+02:00
apply oversampling only to readout for Cartesian trajectories
- - - - -
e7203f43 by Martin Uecker at 2023-06-14T16:45:41+02:00
remove unsed precondition
- - - - -
22b3c07a by Martin Strunz at 2023-06-26T14:53:20+02:00
[misc] rename shadow variables
- - - - -
4d34d264 by Martin Strunz at 2023-07-01T13:11:57+02:00
misc: rename shadow variable
- - - - -
31b3c0b9 by Martin Uecker at 2023-07-01T13:13:16+02:00
fix bounds error
- - - - -
eb71b966 by Martin Uecker at 2023-07-01T13:13:16+02:00
coding style
- - - - -
238fe3ad by Martin Strunz at 2023-07-06T22:35:31+02:00
misc: remove include
- - - - -
ab9ca6d2 by Philip Schaten at 2023-07-07T23:30:25+02:00
add reflectpad
- - - - -
5ee57338 by Philip Schaten at 2023-07-07T23:30:39+02:00
add nlmeans algorithm
- - - - -
ee8f7a4e by Philip Schaten at 2023-07-07T23:30:47+02:00
add nlmeans command
- - - - -
00f38127 by Martin Uecker at 2023-07-07T23:31:45+02:00
remove includes
- - - - -
3ef3cd89 by Martin Uecker at 2023-07-08T16:55:34+02:00
make test conform to the rest
- - - - -
d8c9e045 by Martin Uecker at 2023-07-08T16:55:34+02:00
some spaces
- - - - -
68e200c6 by Martin Uecker at 2023-07-08T16:55:34+02:00
coding style
- - - - -
5f80e7fe by Martin Uecker at 2023-07-08T16:55:34+02:00
format table
- - - - -
ecc25fe5 by Martin Uecker at 2023-07-08T16:55:34+02:00
use correct length
- - - - -
161ca403 by Martin Uecker at 2023-07-08T16:55:34+02:00
includes
- - - - -
8ece26b4 by Martin Uecker at 2023-07-08T16:55:34+02:00
move gauss pdf to flpmath
- - - - -
4fc6b8e3 by Martin Uecker at 2023-07-25T17:27:45+02:00
coding style
- - - - -
cfdccf27 by Martin Uecker at 2023-07-25T17:27:45+02:00
use time dimenion for (undocumented) moving phantom
- - - - -
57b67219 by Martin Uecker at 2023-07-25T17:27:45+02:00
remove unneeded conditional operator in argument to anon/create_cfl
- - - - -
785d7194 by Martin Uecker at 2023-07-25T17:31:37+02:00
fix header for gpu
- - - - -
257121f8 by Moritz Blumenthal at 2023-07-26T17:18:01+02:00
add option to pass diag/fmac tensor without copy
- - - - -
2cbb9a4a by Moritz Blumenthal at 2023-07-26T17:18:01+02:00
implement maps_create by linop_fmac_create
- - - - -
a96b2de9 by Moritz Blumenthal at 2023-07-26T17:18:01+02:00
implement linop_sampling by linop_cdiag
->move to sense/model
- - - - -
bb702f50 by Moritz Blumenthal at 2023-07-26T17:18:01+02:00
add shared-img-dims option to pics
- - - - -
60d12742 by Martin Uecker at 2023-07-26T17:46:23+02:00
remove incorrectly implied nonnull
- - - - -
ad61e16d by Martin Uecker at 2023-07-26T21:41:21+02:00
remove some unsigned
- - - - -
a6d32d93 by Martin Uecker at 2023-07-26T21:41:21+02:00
simplify a bit
- - - - -
6d8d428d by Martin Uecker at 2023-07-26T21:41:21+02:00
make lowmemstack misuse hard errors
- - - - -
8c09b4a9 by Martin Uecker at 2023-07-26T21:41:21+02:00
coding style
- - - - -
526c96ad by Martin Uecker at 2023-07-26T21:41:21+02:00
fix some coding and style errors
- - - - -
e66bc436 by Martin Uecker at 2023-07-26T21:41:21+02:00
remove unsigned
- - - - -
c85c2c6a by Martin Uecker at 2023-07-26T21:41:21+02:00
add assertions to nudft command for unsupported options
- - - - -
eeaf6dc9 by Martin Uecker at 2023-07-28T08:40:42+02:00
remove unneeded header
- - - - -
612d00bf by Martin Uecker at 2023-07-28T11:05:38+02:00
avoid string-op overflow warning
- - - - -
835240c1 by Martin Uecker at 2023-07-28T11:09:31+02:00
transpose
- - - - -
281243d9 by Martin Uecker at 2023-07-28T11:09:44+02:00
noinit adjoint sa
- - - - -
27276fb2 by Martin Uecker at 2023-07-28T11:09:52+02:00
split off interval2
- - - - -
e5af9e7f by Martin Uecker at 2023-07-28T13:46:29+02:00
analyzer false positive
- - - - -
9934e3e5 by Martin Uecker at 2023-07-30T23:45:06+02:00
remove useless casts
- - - - -
1cd0aba5 by Martin Strunz at 2023-07-31T10:13:17+02:00
misc: rename shadow variables
- - - - -
03740d55 by Moritz Blumenthal at 2023-07-31T14:29:55+02:00
variable batch size support in tensorflow
- - - - -
e21baf53 by Moritz Blumenthal at 2023-07-31T14:29:55+02:00
support for placeholdere in dimensions
- - - - -
7db20512 by Moritz Blumenthal at 2023-07-31T14:29:55+02:00
load tensorflow graph without derivatives
- - - - -
0d0a938b by Moritz Blumenthal at 2023-07-31T14:29:55+02:00
support for real tf arguments
- - - - -
2a945224 by Moritz Blumenthal at 2023-07-31T14:29:55+02:00
tf graph function to print nodes
- - - - -
f67a81e6 by Moritz Blumenthal at 2023-07-31T14:29:56+02:00
add tool to debug and apply TensorFlow graph
- - - - -
0238230d by Moritz Blumenthal at 2023-07-31T14:30:16+02:00
add real rand function
- - - - -
a1637196 by Moritz Blumenthal at 2023-07-31T14:30:18+02:00
add gaussian rand to vecops
- - - - -
52ff15e2 by Moritz Blumenthal at 2023-07-31T14:31:16+02:00
add operator p wrapper for nlops
- - - - -
4ac895a2 by Moritz Blumenthal at 2023-07-31T20:09:58+02:00
provide input and output tensors for tf1 graphs
- - - - -
76c3e0e2 by Moritz Blumenthal at 2023-07-31T20:25:06+02:00
fix legacy scaling for basis psf
- - - - -
235f1268 by Martin Uecker at 2023-07-31T20:25:06+02:00
cleanup
- - - - -
ce42ec76 by Christian Tönnes at 2023-08-03T19:45:13+02:00
Added a guide on compiling bart to WebAssembler
- - - - -
a592cd01 by Christian Tönnes at 2023-08-03T20:05:22+02:00
added small web project for testing
- - - - -
bc08641b by Christian Tönnes at 2023-08-03T20:05:49+02:00
added code to test fft & ifft to the test web page
- - - - -
d35f109c by Christian Tönnes at 2023-08-03T20:05:53+02:00
updated instruction to compile bart to WebAssembly
- - - - -
cb4b23bc by Christian Tönnes at 2023-08-03T20:06:01+02:00
added default values for the two new parameters so a normal build still works.
- - - - -
c1c6aa3a by Christian Tönnes at 2023-08-03T20:06:04+02:00
Added PNG option to Makefile
PNG=0 disables libpng and png output.
- - - - -
35461216 by Christian Tönnes at 2023-08-03T20:06:28+02:00
Added FORTRAN option to Makefile.
- - - - -
b7d13338 by Christian Tönnes at 2023-08-03T20:06:31+02:00
new function to list names of all mem files
- - - - -
93b4cd0d by Christian Tönnes at 2023-08-03T20:06:46+02:00
example web project that provides an interface to the bart command line. Only supports memory files.
- - - - -
c2406d3e by Christian Tönnes at 2023-08-03T20:06:57+02:00
script to build webasm files from static bart.o
- - - - -
c5be2458 by Christian Tönnes at 2023-08-03T20:06:59+02:00
removed binary wasm and autogenerated js files
- - - - -
cc71c2a3 by Christian Tönnes at 2023-08-03T20:07:09+02:00
added bsd license headers to my source files
- - - - -
98c6c3ac by Christian Tönnes at 2023-08-03T20:07:11+02:00
small changes to the documentary and renamed build script for the webasm files
- - - - -
711e7395 by Martin Uecker at 2023-08-03T20:57:55+02:00
coding style
- - - - -
8830bdd5 by Christian Holme at 2023-08-04T18:11:16+02:00
add good debug_backtrace
requires libdw and libunwind, installable as
libdw-dev and libunwind-dev
on debian. For static linking,
libbz2-dev liblzma-dev
and additionally required.
For now, prints backtrace on every error, which is not desirable.
- - - - -
ccfd6eb7 by Moritz Blumenthal at 2023-08-04T18:11:16+02:00
assert with backtrace
- - - - -
d614a386 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
fix memleak
- - - - -
c74a6cfe by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
simplify multiplace
- - - - -
d2756c14 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
debug print backtrace in string
- - - - -
1e0378ed by Moritz Blumenthal at 2023-08-04T18:11:16+02:00
deactivated code for saving backtrace in memcache
- - - - -
e07f931a by Moritz Blumenthal at 2023-08-04T18:11:16+02:00
opt for float vector with var length
- - - - -
6d9240bd by Bernhard Rapp at 2023-08-04T18:11:16+02:00
adds VECN
- - - - -
0f387f67 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
make opt parsing threasafe
- - - - -
0fdb4084 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
make save_commandline threadsafe
- - - - -
22df3720 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
make iofiles threadsafe
- - - - -
e3447767 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
Make fftw plan creation threadsafe
- - - - -
6d9dd082 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
add md_unravel
- - - - -
de8eac77 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
prevent reinitialization of gpu
- - - - -
a3a32e05 by Blumenthal, Moritz at 2023-08-04T18:11:16+02:00
Option to turn off device switching for cuda
- - - - -
5dc62521 by Bernhard Rapp at 2023-08-04T18:11:16+02:00
add generic loop interface to bart
- - - - -
18d10371 by Bernhard Rapp at 2023-08-04T18:11:16+02:00
support openmpi for batch processing
- - - - -
238900ee by Bernhard Rapp at 2023-08-04T18:11:16+02:00
bart loop with omp
- - - - -
f6e84ccc by Moritz Blumenthal at 2023-08-04T18:11:16+02:00
loopig for estdelay
- - - - -
934b6d7e by Moritz Blumenthal at 2023-08-04T18:11:16+02:00
use generic looping for pics looping
- - - - -
e1d94c48 by Martin Uecker at 2023-08-04T20:09:01+02:00
new contributors
- - - - -
c954cff8 by Martin Uecker at 2023-08-04T21:53:14+02:00
some style changes
- - - - -
6358d458 by Martin Uecker at 2023-08-04T21:53:14+02:00
changing perrors to errors
- - - - -
d0cf2017 by Martin Uecker at 2023-08-04T21:53:14+02:00
fix command-line help
- - - - -
e66ccf6a by Martin Uecker at 2023-08-06T14:28:56+02:00
analyzer
- - - - -
a61cdc38 by Martin Uecker at 2023-08-08T09:00:27+02:00
remove unsigned
- - - - -
d132a4bf by Martin Uecker at 2023-08-10T17:02:41+02:00
add warning rotation steps
- - - - -
b3e57a32 by Martin Uecker at 2023-08-10T17:02:41+02:00
remove casts and suppress warning
- - - - -
adf11ca3 by Martin Uecker at 2023-08-12T19:58:17+02:00
fix file input
- - - - -
bbffa751 by Martin Uecker at 2023-08-13T21:59:40+02:00
Revert "apply mask on image, not coil image"
This reverts commit f7faa38427ac371d45b0bc89a1e6ec3183d86b36.
- - - - -
dfa0741c by Martin Uecker at 2023-08-14T15:03:32+02:00
fix for warning
- - - - -
e25c0398 by Martin Uecker at 2023-08-18T17:16:18+02:00
tests for twixread
- - - - -
d679ddf4 by Martin Uecker at 2023-08-18T21:12:07+02:00
missing factor in comment
- - - - -
209c6f88 by Martin Uecker at 2023-08-18T21:12:07+02:00
make update of version more robust
- - - - -
5236a036 by Holme, Hans Christian Martin at 2023-08-21T15:57:21+02:00
gitlab-ci: use "make all" for Build_Tensorflow
Also tensorflow should use "make all", to make sure we can actually build all tools
- - - - -
b056f141 by Christian Holme at 2023-08-22T07:43:22+02:00
gitlab-ci: use "tensorflow_dir" to avoid nameclash with the tool
- - - - -
31adc564 by Martin Uecker at 2023-08-22T12:35:48+02:00
rename some local functions
- - - - -
2dd432f4 by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
add IR LL to mobafit
- - - - -
517b48c0 by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
add basis to mobafit
- - - - -
32bbd58c by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
add batched functions to vec_iter_s
- - - - -
65329cbf by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
add batched (blockdiagonal) conjgrad
- - - - -
9e0132d6 by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
batched conjgrad and gpu for mobafit
- - - - -
1c785f83 by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
add md_zsmin2
- - - - -
6175603f by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
bounds, scale and init in mobafit
- - - - -
772ef763 by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
thresh max
- - - - -
4dcbc321 by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
unify cpu and gpu for div
- - - - -
347d70da by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
init gpu in mobafit
- - - - -
89b935ff by Moritz Blumenthal at 2023-08-22T12:35:48+02:00
add tests for mobafit irlll, basis, gpu
- - - - -
59002378 by Moritz Blumenthal at 2023-08-22T12:47:25+02:00
more efficient zscalar and scalar
- - - - -
d42c5487 by Moritz Blumenthal at 2023-08-22T12:47:25+02:00
write dot kernels for gpu
- - - - -
c909b028 by Moritz Blumenthal at 2023-08-22T12:47:25+02:00
fixup calling tf 2 wrapper forward needs grad_y
- - - - -
c88e50a1 by Moritz Blumenthal at 2023-08-22T12:47:25+02:00
use optimized kernels in computation of std
- - - - -
701e9d16 by Moritz Blumenthal at 2023-08-22T12:49:37+02:00
fixup UB check reduce inner
- - - - -
4c37e906 by Martin Uecker at 2023-08-23T08:49:31+02:00
output version when not reproducible
- - - - -
a769c397 by Christian Holme at 2023-08-23T11:28:49+02:00
version: show tensorflow and MPI support
- - - - -
71c8ca91 by Christian Holme at 2023-08-25T09:35:28+02:00
misc.h: add "do { } while (0)" to assert() redefinition
This prevents a misinterpretation. Using `DEBUG_DWARF=1` without this
`do { } while (0)`-wrapper, an `else` following `assert`
might be compiled as belonging to the `if (!expr)` of `assert`.
For example, this happens in `pocsense.c`.
- - - - -
63eb6694 by Christian Holme at 2023-08-30T15:50:16+02:00
gitlab-ci: make wget less verbose
- - - - -
e5145ea6 by Christian Holme at 2023-08-30T15:50:16+02:00
tests/gpu: relax gpu-gridding test
- - - - -
8263e41a by Christian Holme at 2023-08-30T15:50:16+02:00
gitlab-ci: install bc for some tests
some tests in network.mk use this for calculation
- - - - -
a3c9dc31 by Christian Holme at 2023-08-30T15:51:18+02:00
gitlab-ci: updates for test server
- - - - -
eb047635 by Nick Scholand at 2023-08-30T16:20:38+02:00
fix name
- - - - -
1eda98fb by Nick Scholand at 2023-08-30T16:20:38+02:00
Add tool for special function evaluation with tests
- - - - -
96886bcc by Christian Holme at 2023-09-01T10:12:51+02:00
gitlab-ci: test with custom image
which has most dependencies installed
- - - - -
ee13d0f2 by Martin Uecker at 2023-09-01T13:05:41+02:00
linalg functions
- - - - -
d9f84ea9 by Martin Uecker at 2023-09-01T18:03:38+02:00
complex gaussians
- - - - -
16ef03d8 by Christian Holme at 2023-09-02T16:01:09+02:00
tests/pics.mk: fix test prereqs
- - - - -
4fbdfb81 by Christian Holme at 2023-09-02T16:25:00+02:00
sub: check for bart and not sub
as we call bart later, we need to make sure that it exists
- - - - -
3da9011b by Christian Holme at 2023-09-05T14:28:17+02:00
tests: fix some problems
- add some missing prereqs
- fix typo of TESTS_SLOW
- tests/network.mk: add PYTHONPATH to python calls
- tests/network.mk: remove all temporary files
- - - - -
3e71a983 by Christian Holme at 2023-09-05T14:48:19+02:00
gitlab-ci: tensorflow is now part of the docker image
- - - - -
20986c1d by Christian Holme at 2023-09-05T16:28:22+02:00
make clean: also remove gpu utests
- - - - -
48f960f0 by Christian Holme at 2023-09-05T16:28:22+02:00
calc.c: add cppmap header include
required for the MAP macro
- - - - -
bba38668 by Christian Holme at 2023-09-05T16:28:22+02:00
add BART_ prefix to environment variables
but keep support for old TOOLBOX_PATH and DEBUG_LEVEL
- - - - -
612afdbb by Christian Holme at 2023-09-05T16:28:22+02:00
compile commands into subdir
only bart stays in the top level dir, all commands go into the commands/
subdir
- - - - -
b36e0c2f by Christian Holme at 2023-09-05T16:28:22+02:00
move sub into commands/, add test
and document behavior
- - - - -
06dd44e5 by Christian Holme at 2023-09-05T16:28:22+02:00
gitlab-ci: add commands/ to saved artifacts
- - - - -
6482f46d by Christian Holme at 2023-09-11T13:30:49+02:00
for commands: prefer builtin, then check other locations
additionally, we now check the commands/ subdir at the bart exe
location, even if it is not in BART_TOOLBOX_PATH or TOOLBOX_PATH
- - - - -
7e3f5cc6 by Christian Holme at 2023-09-11T13:30:49+02:00
bart.c: restore bart_exit() calls after error()
If we ever change error to no longer call abort(), they will be very
useful
- - - - -
df85a49f by Christian Holme at 2023-09-11T13:30:49+02:00
gitlab-ci: upload GPU, tensorflow, and LTO artifacts
as the maximum artifact size was increased
- - - - -
87a787ff by Christian Holme at 2023-09-11T13:30:49+02:00
gitlab-ci: better format for GPU tests
no ugly multiline scripts
- - - - -
7318b007 by Christian Holme at 2023-09-11T13:30:49+02:00
Makefile: add MAIN_LIST to command targets
- - - - -
2fef74b3 by Christian Holme at 2023-09-11T13:30:49+02:00
utest.h: output name of failed function
- - - - -
aa2f68b0 by Christian Holme at 2023-09-12T13:10:13+02:00
opts: remove old code
that was #ifdef'd out anyway
- - - - -
f210544a by Christian Holme at 2023-09-12T13:10:23+02:00
fix bart opts parsing
We now stop parsing bart options at the first unknown option.
For this, we need to prepent '+' to the optstr passed to ya_getopt,
which needs a new argument to our `options()`, `bool stop_at_nonopt`.
Additionally, `parse_bart_opts()` now prints the correct usage string
(generated from the bart options), and changes argv and argc to remove
the options it parsed.
- - - - -
1f604621 by Christian Holme at 2023-09-13T14:07:15+02:00
version: output environment-variable style if verbose, add clang version
- - - - -
1f9b11a3 by Christian Holme at 2023-09-13T15:36:41+02:00
version: remove clang version
turns out that clang normally claims to be GCC and defines both __GNUC__
and __VERSION__
- - - - -
410d5fc1 by Christian Holme at 2023-09-13T15:36:41+02:00
Makefile: remove unused BTARGETS variable
probably a leftover from 560cc4c540d92243ad4f353c8f1e6e60ef20245d
- - - - -
8eb5d057 by Christian Holme at 2023-09-13T15:36:41+02:00
Makefile: error on OMP=0 with CUDA=1
as we do not support that configuration
- - - - -
5c6a2449 by Christian Holme at 2023-09-13T15:36:41+02:00
Makefile: remove superfluous quotes
- - - - -
f596c33d by Christian Holme at 2023-09-13T15:37:22+02:00
Makefile: do not add -Wno-vla-parameter for clang
- - - - -
6ca5b8cb by Christian Holme at 2023-09-13T15:38:20+02:00
misc/opts.c: fix array bounds on options()
- - - - -
4b05ee20 by Christian Holme at 2023-09-13T16:32:06+02:00
gitlab-ci: enable OMP for clang
- - - - -
fe63943a by Christian Holme at 2023-09-13T16:32:10+02:00
gitlab-ci: add Clang GPU tests
- - - - -
2ea8d6bf by Martin Uecker at 2023-09-14T17:15:16+02:00
refactor
- - - - -
eb2fb27d by Martin Uecker at 2023-09-14T17:15:16+02:00
bloch / pulse tool
- - - - -
38843b86 by Martin Uecker at 2023-09-14T17:57:35+02:00
bloch mcconnel
- - - - -
1a0bdebb by Nick Scholand at 2023-09-15T13:35:43+02:00
Add SE model with IR option to signal tool with tests
- - - - -
9717a4dd by Nick Scholand at 2023-09-15T13:35:55+02:00
Add warning for deprecated TSE model
- - - - -
11577a6a by Nick Scholand at 2023-09-15T13:37:44+02:00
Set default index in SE model to 1
- - - - -
4707e6c5 by Martin Heide at 2023-09-18T10:15:19+02:00
added modelfit t1
- - - - -
a9a6f46f by Martin Heide at 2023-09-18T10:15:39+02:00
added abs(model) to mobafit
- - - - -
2d6ec2d3 by Martin Heide at 2023-09-18T10:15:42+02:00
add dest for abs(t1fit)
- - - - -
5f6b0691 by Nick Scholand at 2023-09-19T10:18:17+02:00
Replace deprecated model
- - - - -
132b1226 by Martin Uecker at 2023-09-19T13:33:54+02:00
white space
- - - - -
a48af60b by Christian Holme at 2023-09-20T16:27:44+02:00
tests/pics.mk: remove temporary files
- - - - -
179f36a2 by Philip Schaten at 2023-09-21T17:18:09+02:00
Allow OMP=0 CUDA=1 compilation
- - - - -
f1192bc2 by Philip Schaten at 2023-09-21T17:18:09+02:00
reduce number of calls to cuda_get_max_threads
- - - - -
a63cc546 by Philip Schaten at 2023-09-21T17:18:09+02:00
early out in memcache lookup if pointer is not in range
- - - - -
ccc5b519 by Martin Uecker at 2023-09-22T21:54:04+02:00
add bounds-strict for UBSan
- - - - -
70c64867 by Christian Holme at 2023-09-25T12:03:22+02:00
toimg: use construct_filename from view
For consitency between toimg and view/cfl2png, it makes more sense to
have this function in bart
Now, if `toimg -D` is used, the filenames output from toimg and
view/cfl2png match
This adds the improvements added in view commit
1e8abf212036251136ac39c75f7b8e9318b2b497
- - - - -
774527b7 by Christian Holme at 2023-09-25T12:04:27+02:00
toimg: rename loop_flags to im_flags, as they are the image flags
- - - - -
8af06a2c by Martin Uecker at 2023-09-30T12:15:14+02:00
coding style
- - - - -
3ce0ff0e by Martin Uecker at 2023-09-30T12:15:14+02:00
fix uint (closes #318)
- - - - -
31906897 by Martin Uecker at 2023-09-30T12:15:14+02:00
include stdio for FILE (closes: #317)
- - - - -
8dae4631 by Christian Holme at 2023-09-30T15:12:10+02:00
bart.c: fix option handling
calling bart with a global bart option but without a command
(`bart -l 7`) would fail, as `argv[0]` is `NULL` after advacing argv.
Therefore, we catch that case and print the available commands.
- - - - -
6eadb63e by Moritz Blumenthal at 2023-10-02T11:40:56+02:00
fixup gpuops bat
- - - - -
07a3760c by Moritz Blumenthal at 2023-10-02T11:40:57+02:00
fix zero copy case
- - - - -
de404b79 by Martin Uecker at 2023-10-04T10:07:38+02:00
minor
- - - - -
e3766eba by Martin Uecker at 2023-10-04T10:07:38+02:00
fix bounds error
- - - - -
07d0118f by Bernhard Rapp at 2023-10-04T10:09:01+02:00
cleanup mpi, shared filesystems, integration tests
- - - - -
332e3e17 by Moritz Blumenthal at 2023-10-04T10:09:01+02:00
mmio updates
- - - - -
aa77268a by Moritz Blumenthal at 2023-10-04T10:09:01+02:00
mpi prevent reinitialization of GPU
- - - - -
0cd9484c by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds md_select_strides
- - - - -
220ac50a by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
Add md_ravel_index
- - - - -
42db8b7e by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
adds virtual pointer
- - - - -
d9f22917 by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
Add gpu support to vptr
- - - - -
16a33ed0 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
extends MPI API
- - - - -
795e1f06 by Moritz Blumenthal at 2023-10-04T10:09:01+02:00
update debugging for mpi
- - - - -
01fc6ed0 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds MPI to multind
- - - - -
e2dc5e23 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds MPI support to flpmath
- - - - -
025891a9 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
sameplace for scalar_ops
- - - - -
131c6195 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds MPI to fft
- - - - -
bb42d2e0 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds MPI unit tests
- - - - -
e1cd76d5 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds lin(op) MPI APIs
- - - - -
cc62e03f by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
add openmpi for gridding
- - - - -
3ae7e812 by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
update multiplace for vptr
- - - - -
379261b2 by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
estimate scaling using md functions
- - - - -
d4e305a4 by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
pics weighting in forward op
- - - - -
f1e972be by Bernhard Rapp at 2023-10-04T10:09:01+02:00
sameplace for estimate pattern
- - - - -
7fe6ad30 by Bernhard Rapp at 2023-10-04T10:09:01+02:00
adds MPI to pics
- - - - -
610a370e by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
move mpi abort to mpi ops
- - - - -
f1058e2a by Blumenthal, Moritz at 2023-10-04T10:09:01+02:00
only initialize mpi when requested
- - - - -
86219918 by Martin Uecker at 2023-10-04T10:15:40+02:00
minor warning fixes
- - - - -
2d626d20 by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
adds averaging losses for nn_stack_multigpu_F
- - - - -
4c8993aa by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
mpiops add vector reduction
- - - - -
508d7ae6 by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
replace multigpu stack by openmpi stack
- - - - -
2ae69139 by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
mpi support for reconet
- - - - -
41ef2e87 by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
Print warning on initializationn error
- - - - -
d9619e1d by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
simplify gpuops (multigpu via mpi)
- - - - -
eed48b37 by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
unify gpu initialization
- - - - -
ac8a7a96 by Moritz Blumenthal at 2023-10-04T22:35:54+02:00
update doc for mpi
- - - - -
ed6efe02 by Moritz Blumenthal at 2023-10-05T10:13:21+02:00
fix zdiv_reg on gpu
- - - - -
22120e7e by Moritz Blumenthal at 2023-10-05T10:13:21+02:00
fixup estdelay
- - - - -
13548a38 by Blumenthal, Moritz at 2023-10-05T10:13:21+02:00
fix warning
- - - - -
499d5954 by Moritz Blumenthal at 2023-10-05T10:13:21+02:00
fix tf checkpoint on gpu
- - - - -
5ce463a6 by Moritz Blumenthal at 2023-10-05T10:13:21+02:00
fix tf1 tests not deleting mapping
- - - - -
7cd34d7d by Blumenthal, Moritz at 2023-10-05T11:56:45+02:00
fix stream selection
- - - - -
f7862a95 by Blumenthal, Moritz at 2023-10-05T15:14:57+02:00
fix warning
- - - - -
7fa17ebd by Martin Juschitz at 2023-10-07T15:43:33+02:00
add bloch_mcconnell_ode
- - - - -
0b970939 by Martin Juschitz at 2023-10-07T15:43:37+02:00
Add rectangular pulse with test
- - - - -
88d72a2e by Nick Scholand at 2023-10-07T15:43:37+02:00
Simplify pulse type handling
- - - - -
de8a9061 by Nick Scholand at 2023-10-07T15:43:37+02:00
Prepare simulation for Bloch-McConnell integration
- - - - -
d77025e4 by Nick Scholand at 2023-10-07T15:43:37+02:00
Fix STM size and make relaxation2 accessible
- - - - -
06579736 by Martin Juschitz at 2023-10-07T15:43:37+02:00
Change Th -> m0
- - - - -
bb614604 by Martin Juschitz at 2023-10-07T15:43:37+02:00
Update k-handling for Bloch-McConnell simulations
- - - - -
532f4e59 by Martin Juschitz at 2023-10-07T15:43:37+02:00
Add Jacobians of direct sensitivity analysis of Bloch-McConnell with test
- - - - -
e8b7cd74 by Martin Juschitz at 2023-10-07T15:43:37+02:00
Add direct sensitivity analysis of Bloch-McConnell with test
- - - - -
c20810de by Martin Juschitz at 2023-10-07T15:43:37+02:00
Add Bloch-McConnell support to simulation backend with tests
- - - - -
3e62f131 by Martin Juschitz at 2023-10-07T15:43:37+02:00
Add Bloch-McConnell to sim interface
- - - - -
bd1f796d by Martin Juschitz at 2023-10-07T15:43:37+02:00
Add BMC CEST test
- - - - -
8f7571ab by Martin Uecker at 2023-10-07T15:43:37+02:00
white space and programming style
- - - - -
daca5f92 by Martin Heide at 2023-10-10T13:41:58+02:00
added ellipsoid simulation to phantom tool
- - - - -
47ded36d by Martin Heide at 2023-10-10T14:05:04+02:00
added test for ellipsoid phantom simulation
- - - - -
7a9578bf by Martin Uecker at 2023-10-12T20:56:16+02:00
coding style
- - - - -
30cb5a93 by Moritz Blumenthal at 2023-10-13T13:49:04+02:00
sync cuda streams before cuda aware mpi ops
- - - - -
dd88d078 by Blumenthal, Moritz at 2023-10-17T15:39:00+02:00
copy data back to gpu
- - - - -
d1141683 by Blumenthal, Moritz at 2023-10-19T13:39:11+02:00
fixup unmap
- - - - -
ca8226e2 by Martin Uecker at 2023-10-25T21:56:38+02:00
style
- - - - -
22e4a100 by Martin Uecker at 2023-10-25T21:56:38+02:00
fixes from VLA-enhanced object size pass
- - - - -
d072aebe by Nick Scholand at 2023-10-25T21:57:37+02:00
Tool to calculate PSF of given trajectory
- - - - -
a77d2179 by Martin Uecker at 2023-10-25T21:57:37+02:00
rational approximation of golden angles
- - - - -
e90ff6e4 by Nick Scholand at 2023-10-25T21:57:37+02:00
Unit test and fix for Fibonacci series
- - - - -
4c3a4020 by Nick Scholand at 2023-10-25T21:57:37+02:00
2 Pi base option for GA
- - - - -
4ca760ab by Nick Scholand at 2023-10-25T21:57:37+02:00
Separate functions for RAGA sampling, UI checks and tests
- - - - -
0fb99b29 by Nick Scholand at 2023-10-25T21:57:37+02:00
Add temporal reordering of RAGA sampled data to twixread
- - - - -
88c30280 by Nick Scholand at 2023-10-25T21:57:37+02:00
Recover GA index from RAGA increment
- - - - -
9950087d by Nick Scholand at 2023-10-25T21:57:37+02:00
Binning option in twixread for RAGA sampled data
- - - - -
df2c3422 by Nick Scholand at 2023-10-25T21:57:37+02:00
style
- - - - -
f9239b66 by Martin Uecker at 2023-10-25T22:29:32+02:00
parameter consistency
- - - - -
bc5ba1ae by Martin Uecker at 2023-10-25T22:37:36+02:00
do not recover from UBSan error
- - - - -
d46adcd9 by Jon Tamir at 2023-11-01T21:20:04+01:00
set DEBUG_LEVEL for tests that read from stdout
- - - - -
63c4e05e by Jon Tamir at 2023-11-01T21:20:27+01:00
relax tolerance for pics test (Mac ARM)
- - - - -
04870197 by Martin Uecker at 2023-11-02T08:27:22+01:00
remove debug prints
- - - - -
240b8e7a by Martin Uecker at 2023-11-02T08:27:22+01:00
incorrect static
- - - - -
2bde1143 by Martin Uecker at 2023-11-02T08:27:22+01:00
unsigned -> int
- - - - -
3c68f72c by Martin Uecker at 2023-11-02T16:32:28+01:00
cleanup
- - - - -
fbf58745 by Martin Uecker at 2023-11-02T19:21:34+01:00
update references
- - - - -
66cbe8b2 by Martin Uecker at 2023-11-04T18:00:42+01:00
check GCC versions, use -std=gnu17
- - - - -
cf7e9c8c by Martin Uecker at 2023-11-04T20:19:58+01:00
Revert "Fix STM size and make relaxation2 accessible"
This reverts commit d77025e4584f0a8f6d254cd6ed6160d1aff484a1.
- - - - -
f9b27e0f by Martin Uecker at 2023-11-04T20:19:58+01:00
minor corrections
- - - - -
16fe4bc0 by Martin Uecker at 2023-11-06T07:39:16+01:00
dependencies
- - - - -
f335950f by Jon Tamir at 2023-11-10T18:15:06+01:00
Update Mac default gcc version and build instructions
- - - - -
6c4ce8d5 by Moritz Blumenthal at 2023-11-15T15:55:56+01:00
respect umask in file creation
- - - - -
fd6ccc1f by Martin Juschitz at 2023-11-19T15:02:13+01:00
Fix potential divide by 0
- - - - -
8ca18499 by Martin Juschitz at 2023-11-19T15:02:28+01:00
Fix order in simulation call and derivatives
- - - - -
60183c7c by Martin Juschitz at 2023-11-19T15:02:31+01:00
Fix maps_dim reset
- - - - -
fcea7c09 by Martin Uecker at 2023-11-19T15:03:15+01:00
whitespace
- - - - -
77130222 by Martin Uecker at 2023-11-19T21:58:24+01:00
rework pulses
- - - - -
14cb8b63 by Martin Uecker at 2023-11-19T21:58:24+01:00
set defaults
- - - - -
589ffaee by Martin Uecker at 2023-11-19T21:58:24+01:00
remove rep_counter
- - - - -
2f2a5f6f by Martin Uecker at 2023-11-19T21:58:24+01:00
remove spin_counter
- - - - -
be5cc8e8 by Martin Uecker at 2023-11-19T21:58:24+01:00
remove unused t
- - - - -
c10e199e by Martin Uecker at 2023-11-19T21:58:24+01:00
remove w1
- - - - -
f7ae2804 by Martin Uecker at 2023-11-19T21:58:24+01:00
remove global state
- - - - -
df866cde by Martin Uecker at 2023-11-19T21:58:24+01:00
add vecf_copy
- - - - -
d1a43631 by Martin Uecker at 2023-11-19T21:58:24+01:00
simplify
- - - - -
dcbb3452 by Martin Uecker at 2023-11-19T21:58:24+01:00
simplify
- - - - -
f4c4f1bf by Martin Uecker at 2023-11-19T21:58:24+01:00
simplify
- - - - -
1363b524 by Martin Uecker at 2023-11-20T10:37:51+01:00
clang workarounds
- - - - -
cbb6016d by Martin Uecker at 2023-11-24T14:19:14+01:00
early model-based literature
- - - - -
49bd1328 by Martin Uecker at 2023-11-30T18:10:40+01:00
analyzer fix
- - - - -
672a840f by Martin Uecker at 2023-12-06T08:30:29+01:00
bump version
- - - - -
e486669e by Martin Uecker at 2023-12-07T10:21:37+01:00
Merge tag 'v0.9.00'
version 0.9.00
Changes:
- new tools: nlmeans, tensorflow, nufftbase, calc, psf
- calc: access some mathematical functions from the command-line
- nlmean: non-local means filter
- tensorflow: load and apply Tensorflow graph
- nufftbase: compute Fourier transform of basis used in nuFFT
- psf: compute PSF
- nufft: various new options to control behavior
- nufft: basis and weights
- nufft: zero and low memory modes
- nufft: use -x for image dimensions
- sim: Bloch-McConnell simulations
- traj: rotation of Cartesian trajectories
- traj: RAGA trajectories
- reconet: MPI support
- reconet: tensorflow graph support
- reconet: export graph for visualization
- pics: mpi and multi-GPU support
- pics: add shared-img-dims option to pics
- pics: add option to specify FISTA
- pics: preconditioning
- nlinv: add option to specifiy non-Cart image size
- phantom: BRAIN geometry
- phantom: 64ch sensitivities
- phantom: ellipsoid simulation
- phantom: support for generic geometries
- mobafit: add basis to mobafit
- mobafit: add IR and IR LL to mobafit
- mobafit: magnitude fitting
- mobafit: scaling and initial values
- mobafit: minimum and maximum bounds
- mobafit: GPU support
- nrmse: add option for scientific notation
- moba: Export and ksp coils in moba
- moba: Various new options
- estdelay: support for looping
- version: show tensorflow and MPI support
- signal: Spin-Echo signal model
- scripts/shell: kspace_precond.sh: Script to compute k-space preconditioner
- scripts/python: bart_tf.py: Export tensorflow graphs for use with BART
- scripts/python: bart.py: Print stdout and stderr output from bart call
- scripts/python: splines_from_svg.py: extract spline control points from SVG files
- library/geom: Brain geometry
- library/misc: better backtraces
- library/num: md_ravel_index, md_select_strides, md_zsmin2, md_unravel
- library/num: library for complex-valued Gaussians
- library/num: cuda kernels: zdot, zfsq2, zfftmod_3d, zsmin, compress, decompress
- library/num: cuda batch kernels: xpay, axpy, dot_bat
- library/num: cuda kernels: (z)add_unfold, (z)mul(c)_unfold, (z)fmac(c)_unfold
- library/num: md_nlmeans(2), md_nlmeans_distance(2)
- library/num: library for MPI sync, broadcast, reductions
- library/num: double precision bessel function
- library/num: special functions: sinc, sincf, jinc
- library/num: library for virtual pointers
- library/num: time-ordered matrix exponentials: mat_to_exp
- library/num: functions for adjoint sensitivity analysis
- library/num: functions for quadratures: trapezoidal, simpson
- library/num: configure fftw wisdom/measure via environment variable
- library/simu: Bloch McConnell simulations
- library/simu: Spin Echo signal model
- library/linops: linop: flip
- library/nlops: nlops: zphsr and zabs
- library/nn: nlops: ardioid, siglog, igaussian
- library/network: tensorflow networks
- library/noncart: nufft inverse gridding on GPU
- library/noncart: RAGA trajectories
- library/iter: weighted least-squares prox function
- library/iter: batched conjgrad
- library/iter: multiple regularizations for primal-dual
- general: webasm support
- general: BART_ prefix to environment variables
- general: generic loop interface to bart
- general: basic MPI support
- general: OpenMPI and GPU for gridding
- general: zero-overhead mode for nuFFT
- general: performance improvements
- general: many GPU improvements, unified GPU initialization
- built: FORTRAN and PNG options
- built: LTO integration test
- built: OMP=0 CUDA=1 compilation
- built: more unit and integration tests
- many other bug fixes and improvements
- - - - -
89af9e06 by Martin Uecker at 2023-12-07T10:21:55+01:00
rebase patches (and drop some)
- - - - -
6fabdd61 by Martin Uecker at 2023-12-07T10:22:06+01:00
udpate copyrights
- - - - -
d45c5d9b by Martin Uecker at 2023-12-07T10:23:41+01:00
changelog
- - - - -
30 changed files:
- .gitlab-ci.yml
- ACKNOWLEDGEMENTS
- Makefile
- README
- README.md
- + build_webasm.sh
- + commands/.gitignore
- + commands/sub
- debian/changelog
- debian/copyright
- debian/patches/0001-remove-empty-directory.patch
- debian/patches/0002-deactivate-ode-unit-tests.patch
- − debian/patches/0004-workaround-for-compiler-bug.patch
- debian/patches/0005-do-not-update-doc-commands.txt-automatically.patch
- debian/patches/0006-on-Deban-never-use-git-version.patch
- debian/patches/0007-add-CXXFLAGS-to-nvcc-to-fix-C-build-issue.patch
- − debian/patches/0008-relax-a-failing-test.patch
- − debian/patches/0009-backport-specfun-fixes.patch
- debian/patches/0010-turn-off-one-test-on-i386.patch
- debian/patches/series
- doc/building.txt
- + doc/distributed-architectures.txt
- doc/references.txt
- + doc/tensorflow.txt
- + doc/webasm.txt
- matlab/get_bart_path.m
- python/bart.py
- + python/bart_tf.py
- + python/splines_from_svg.py
- rules/update-version.sh
The diff was not included because it is too large.
View it on GitLab: https://salsa.debian.org/med-team/bart/-/compare/8a18c5bce0a9cefcfa0bc85801a51dfacc45f1d9...d45c5d9bb7c369c1633e92b1fbb8538b69296fc7
--
View it on GitLab: https://salsa.debian.org/med-team/bart/-/compare/8a18c5bce0a9cefcfa0bc85801a51dfacc45f1d9...d45c5d9bb7c369c1633e92b1fbb8538b69296fc7
You're receiving this email because of your account on salsa.debian.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-med-commit/attachments/20231207/a63b08ed/attachment-0001.htm>
More information about the debian-med-commit
mailing list